sign in sign up
<<
Home , Coastal erosion, Progradation

OF THE UNITED STATES SHORELINE

ORRIN H. PILKEY, JR AND E. ROBERT THIELER Duke Universic. Depunmenf of . Old Chemist? Building. Durham. North Crtrolina 177015

ABSTRACT: Over 75 percent of the United States ocean shoreline is eroding (retreating landward). Shoreline progradation. where oc- curring, is generally assumed to be a temporary phenomenon. When affecting a developed area, shoreline retreat is usually termed erosion. but considerable confusion remains over the use of this term. Retreat and progradation refer to a change in shoreline position. whereas erosion and accretion refer to volumetric changes in the subaerial . As used in this paper. however. erosion refers to any form of shoreline retreat. consistent with common usage. Coastal erosion is a fundamental and widespread process on U.S. and world shorelines. Evidence. particularly on banier- . indicates that in the past few decades or millenia. erosion may have become a more widespread process. Possible causes of this change include the effect of humans, shoreface steepening. or an increase in the rate of eustatic sea-level rise. Mechanisms responsible for shoreline erosion arc highly variable. both temporally and geographically. In addition, our understanding of shoreline -transport dynamics is incomplete. Consequently, we are presently unable to predict accurately future shoreline- retreat rates related to continued sea-level rise. The Bmun Rule. for example. predicts little shoreline retreat relative to using, as a pdictive tool, the slope of the land surface over which sea level is expected to rise.

WHAT IS COASTAL EROSION? a variety of reference features. measurement techniques, and Defining erosion in the coastal environment is difficult. rate-of-ihange calculations. Thus, whereas CEIS represents Various terms are used to describe the process, including the best available data for the U.S. as a whole, much work drowning, coastal erosion, shoreline erosion, beach ero- is needed to document regional and local erosion rates ac- sion. shoreline retreat, beach retreat, and shoreline reces- curately. sion. Strictly speaking, retreat and progradation refer to a When examined in detail, such as on a state-by-state ba- change in shoreline position; erosion and accretion refer to sis, the actual percentage of eroding shoreline is generally volumetric changes in the subaerial beach (Wood and oth- much higher than the coastwide averages shown in Figure ers, 1988; Oertel and others, 1989). The term beach ero- 1 suggest. This discrepancy is likely due to the large amount sion, however, is deeply ingrained in the public lexicon to of shoreline for which "no data" exist and the coarse res- refer to any form of shoreline retreat (usually the change olution of CEIS. Dolan and others (1990b). for example, in position of the high- line); we shall here use it in the report that 78 percent of the Delaware shoreline is eroding; same broad sense as does the general public. Maryland, 94 percent; Virginia, 72 percent; and North Car- When dealing with shorelines and , it is crucial olina, 72 percent. Griggs (1987a) states that approximately to distinksh between erosion and an erosion ~roblem.Many 86 percent of the California shoreline is eroding. kilometgrs of undeveloped U.S. shoreline & eroding (re- treating landward) and, as a rule, such locations are not WHY UBIQUITOUS EROSION? considered to be societal problems. Erosion and retreat of the shoreline are, in fact, critically important processes in Coastal erosion is a fundamental and widespread process the evolution of Holocene coastal . , la- not only on U.S. but also world shorelines. There is some goons, spits, banier , sea cliffs, and many other coastal evidence, however, that such has not always been the case. landforms all owe their shape and depositional patterns to Examination of the U.S. barrier-island shoreline, for ex- erosion, at least in part. It is only when humans interfere ample, indicates that fundamental changes in depositional with or get in the way of shoreline erosion that it is rec- patterns have recently taken place. Many barrier islands are ognized as a problem. A broad discussion of the U.S. coastal- regressive, indicating that a seaward accretion or widening erosion problem, along with methodology, political impli- of the island occurred over the last 2 to 4 ka. Within the cations, and costs for mitigation is furnished by the Na- last several hundred years, however, a profound change has tional Research Council (1990). taken place and shoreline retreat has replaced shoreline pro- gradation. In other words, barrier islands are becoming in- U.S. SHORELINE-EROSION RATES creasingly transgressive. Most regressive banier islands (e.g., Erosion rates for the contiguous U.S. shoreline, based Bogue Banks, North Carolina, and Galveston Island, Texas) on data in the Coastal Erosion Information System (CEIS; are now eroding on both the ocean and sides. May and others, 1982), have been compiled by May and Possible explanations for increased worldwide erosion others (1983) and Dolan and others (1985; Tables 1 and 2; include the effects of humans-shoreline stabilization and Fig. 1). CEIS includes shoreline-change data for the At- damming; supply reduction-shoreface steep- lantic, of Mexico, Pacific and Great Lakes coasts, as ening; and a recent increase in the rate of eustatic sea-level well as for major bays and estuaries. The data in CEIS are rise. However appealing, each of these possibilities leaves drawn from a variety of sources, including published re- certain questions unanswered. For example, humans cer- , historical shoreline-change maps, field surveys, and tainly contribute to coastal erosion in some locations, but aerial-photograph analyses. However, the lack of a stan- completely undeveloped areas such as the bamer-island dard method among coastal scientists for analyzing - system on the Pacific of Colombia are also eroding line changes has resulted in the inclusion of data utilizing at high rates.

Quaternary Coasts of the United States: Muine and Lacusvine Systems. SEPM Special Publication No. 48 Copyright 0 1992. SEPM (Society for Sedimcnfaiy Geology). ISBN 0-918985-98-6 ORRIN H. PILKEY. JR AND E. ROBERT THIELER

TABLEI .-SHORELINE RATES OF CHANGE FOR REGIONS AND STATES

Mean SUndmd Region trn/y~l Deviation ~mgel Atlantic Coast -0.8 3.2 25.5/-24.6 Maine -0.4 0.6 1.91-0.5 New Hampshire -0.5 - -0.5/-0.5 Massachu~ns -0.9 1.9 4.5/-4.5 19 Rho& Islad -0.5 0. I -0.3/-0.7 I Aa New York 0.1 3.2 18.8/-2.2 a- m New Jersey zs.S/- 15.0 Delaware 5.0/-2.3 Atlantic Coast Gulf of Mexico Maryland 1.3/-8.8 Virginia 0.9/-24.6 Nmh Carolina 9.4/-6.0 South Carolina 5.9/-17.7 Erosion Rate (m/lvr) Georgia 5.0/-4.0 Florida 5.0/-2.9 Gulf of Mexico 8.8/- 15.3 Florida 8.8/-4.5 1 1.1-2 Alabama 0.8/-3.1 Mississippi 0.6/-6.4 Louisiana 3.4/- 15.3 Texas 0.8/-5.0 Pacific Coast 10.0/-4.2 El NoDw California 10.0/-4.2 Oregon -0.1 1.4 SO/-5.0 FIG. 1.-Pie charts of erosion rates on the U.S. open-ocean coastline. -0.5 2.2 Washington 5'0/-3'9 The large amount of shoreline classified as "no data" likely results in 'Positive values indicate accretion; negative values indicate erosion. (After Dolan an artificially low figure for the total percentage of eroding shoreline (see and others. 1985.) text for discussion). (After Dolan and others, 1990a).

TABU 2.-SHORELINE RATES OF CHANGE FOR COASTAL On barrier-island coasts, erosion rates must be examined nPEs in the context of the total barrier-island system. The con- development of transgressive and regressive -

-- riers along the Texas coast, for example, is related to the Burier islands longshore distribution of sediment sources and sinks (Mor- Atlantic Coast -0.8 3.4 25.5/-24.6 Maine-New Yorlr -0.3 2.6 4.51-1.5 ton, 1979; Numrnedal, 1983). New York-North Carolina -1.5 4.5 25.5/-24.6 Shoreline erosion on the Texas coast predominates at del- Nonh Carolina-Florida -0.4 2.6 9.4/- 17.7 taic , such as the Rio Grande and Brazos deltas. Gulf of Mexico -0.6 1.5 8.8/-4.3 Florida-Louisana -0.5 1.7 8.8/-4.3 In the central portion of each embayment between the del- Louisiana-Texas -0.8 1.2 0.8/-3.5 taic headlands, accretion or at least relatively slow erosion Sud beaches Atlantic Coast -1.0 I .O 2.0/-4.5 occurs (Morton, 1979). The eroding deltaic headlands are Maine-Massachusetts -0.7 0.5 -0.5/-2.5 the source of sand for the accreting shoreline between river Massachusetts-New Jersey -1.3 1.3 2.0/-4.5 mouths. Gulf of Mexico -0.4 1.6 8.8/-4.5 Pacifif Coast -0.3 1.O 0.7/-4.2 Evolution of the Louisiana coast is similarly affected by Sad beaches with -0.3 1.9 lO.O/-5.0 regional-scale processes. Penland and others (1985) dem- Pocket beaches onstrated a cyclic relation between Mississippi Delta lobe- Atlantic Coa5t -0.5 - -0.5/-0.5 Prifu Coast -0.2 1.1 5.01- I. I switching events and barrier development and destruction. Dcltsr -2.5 3.5 8.8/- 15.3 Extremely rapid shoreline-erosion rates prevail no matter Mud flur Gulf of Mexico -1.9 2.2 3.4/-8.1 what the stage of barrier development in this cycle. The Flaidr -0.3 0.9 1.5/- 1.5 high overall erosion rates are fundamentally due to subsi- Louisiana-Texas -2.1 2.2 3.4/-8.1 dence-induced rapid rates of sea-level rise (0.50-1.00 cm/ Rock rhaclincr Atlantic Coast 1.O 1.2 1.91-4.5 yr; Penland and others, 1985). However, the rates of shore- Pa~ifuCoast -0.5 - -O.S/-0.5 line recession, which vary from 5 to 15 m/yr (Penland and 'Positive values indicate accretion; negative valuer indicate erosion. (After Dolan others. 1985), depend partly on the age of the individual md ahm. 1985.) barrier-island chains. Beaches on young barrier chains on the Mississippi Delta (e.g., Isles Dernieres) have relatively lower erosion rates because the sand supply (the eroding deltaic lobe) is close to sea level. As the lobe subsides, the LOCAL FACTORS AFFECTING EROSION sand source is effectively removed and shoreline-erosion On all types of sandy or unconsolidated coasts, beach rates increase. erosion can be considered to be controlled by a dynamic Antecedent geology has also been shown to affect barrier equilibrium involving three major components: sediment development and hence shoreline-erosion rates. Barrier is- supply, wave and tidal energy, and the position of, or changes lands grounded on antecedent topographic features (e.g., in, sea level. Evans and others, 1985) may, at least temporarily, expe- U. S. SHORELINE EROSION rience reduced erosion rates. In addition, the Holocene/ Trenhaile (1987). Factors controlling erosion on the Caii- Pleistocene underlying the barrier islands may fornia coast are addressed by Griggs and Savoy (1985). play a major role in determining shoreline-erosion rates. The degree of retreat in California depends on (1) Thick sequences of lagoonal or salt- mud are respon- orientation or exposure, (2) local wave climate, and (3) the sible for the rapid shoreline-erosion rates of some islands physical properties of the cliff. Important rock properties along the Delmarva (Demarest and Leatherman, include hardness, consolidation, and the presence or ab- 1985). The high beach-erosion rates are due to a combi- sence of internal weaknesses such as joints, fractures, and nation of (1) ease of erosion of a muddy shoreface, and (2) faults. Crystalline rocks such as granite tend to erode ir- a lack of sand contribution from the eroding shoreface (De- regularly and very slowly, producing a highly irregular marest and Leatherman, 1985). coastline. Sedimentary rock, broadly speaking, tends to erode Waves associated with extratropical and tropical storms somewhat more regularly and rapidly, producing relatively cause the most visible and obvious shoreline erosion, but straight coasts. often storm-caused erosion is substantially "repaired" by Both marine and nonmarine processes are involved in cliff post-storm onshore and longshore . retreat (Noms, 1990; Emery and Kuhn, 1982). Marine ero- Shoreline retreat may occur under both fairweather and storm sion includes direct wave attack, a variety of chemical-so- conditions. The relative propoxtion of fairweather and storm lution processes, and flaking by salt-crystal expansion. retreat is not well documented, but it probably varies widely Nonmarine cliff erosion is accomplished by chemical and from beach to beach. In addition, different shorelines are mechanical processes usually involving rain water. adjusted to different wave climates. New and Mid- Groundwater seepage along contacts and joints is often a Atlantic shorelines, for example, frequently retreat on an critical part of cliff retreat. Along the California coast, the annual basis in response to "northeaster" storms. Erosion effects of seepage have been increased by lawn watering of Florida's Gulf of Mexico beaches, on the other hand, is and septic-tank emplacement (Noms, 1990). most affected by hurricanes, often spaced decades apart. Mass movement in the form of slumps and various types The role of humans in causing coastal retreat ranges in of landslides is frequently the major process of retreat and scope from global to local. Global production of excess C02 is responsible for the very uneven spatial and temporal re- is widely assumed to be leading to "greenhouse effectv- treat rates of cliffs. Emery and Kuhn (1982) present a clas- related sea-level rise and accelerated erosion. Locally, the sification of coastal cliffs based in part on the relative im- damming of is cutting off a major source of sand for portance of marine and nonmarine processes. many beaches, which accelerates erosion. This problem is According to Griggs and Savoy (1985), erosion rates along acute for California beaches (Griggs, 1987b). pre- the rocky coast of California are highly variable, ranging vention on rivers by the armoring of banks and from negligible to 3 to 4 m/yr. Typically, shoreline-retreat construction of artificial accelerates erosion by in- rates are on the order of 10 to 20 cm/yr. Griggs and Savoy hibiting vertial accretion of river deltas and further reducing also note that the highly episodic nature of erosion rates here is strongly dependent on storminess. Frequent storms beach sediment supply. ' The myriad sea walls, breakwaters, groins, and jetties translate to high-erosion rates, and the degree of storminess that line developed shorelines divert offshore, slow down, itself is episodic. trap, and otherwise reduce the regional beach sediment sup plied by longshore currents, and thereby increase erosion PREDICTING FUTURE SHORELINE CHANGES rates. construction along formerly glaciated coasts One of the most critical research problems in coastal ge- has cut off the supply of sand normally contributed to the ology today is predicting the effect of sea-level rise on beaches by eroding bluffs. This phenomenon has led the shoreline-retreat rates. Coastal communities need accurate state governments of Massachusetts and Maine to prohibit predictions of shoreline position on a time frame of de- the construction of sea walls in front of bluffs. cades, an impossibility at present. A spectacular example of the role of jetties in affecting Were sea level to rise catastrophically 10 m in the next beach sand supply, and also a spectacular example of beach year, there is no question that the shoreline would be inland erosion and barrier-island migration, are afforded by the at what is now the 10-m contour. Predicting shoreline be- northern end of Assateague Island, Maryland (Leatherman, havior at a slower rate of sea-level rise, however, is much 1984). After the Ocean opened during a hurricane more difficult. The problem is complex, and standard en- in 1933, jetties were constructed to stabilize the inlet. Since gineering models such as (Bruun, 1962) are that time, sand that would have been transported across the too rigid and are based on too many narrow assumptions inlet to Assateague has been accumulating updrift of the to have wide application (Kraft and others, 1987; Orford, jetties in front of Ocean City. or offshore at the end of 1987; SCOR Working Group 89, 1991). the jetties. This diversion of sediment has caused the As- Methods of predicting future shoreline changes include sateague shoreline to retreat and the island itself to migrate extrapolation of present erosion rates, application of pre- landward. Today, the open-ocean of Assateague dictive models, and use of numerical models. Present ero- Island is landward of the island's 1933 lagoon shoreline. sion rates are often used by states to determine building The causes and mechanics of erosion are more complex setbacks (e.g., in North Carolina and SouthCarolina). The on rocky coasts than on unconsolidated coasts. Processes basic problem with this method, however, is that recent involved in the erosion of rocky coasts are summarized in trends and their underlying causes may not in any way re- ORRIN H. PILKEY, JR AND E. ROBERT THIELER

flect the future. Dolan and others (1991) provide a review BUN. R.. FENSTER. M.. AND HUE. S.. 1991. Temporal analysis of of the methods used to calculate rates of shoreline change shoreline recession and accretion: Journal of Coastal Research, v. 7. p. 723-744. using historical data. DOLAN,R.. TROSSBACH.S.. AND BUCKLEY.M.. 1990b. New shoreline A number of fairly simple predictive models are used to erosion data for the mid-Atlantic coast: Journal of Coastal Research. project future shoreline positions, including the Bruun Rule v. 6, p. 471-477. (Bruun, 1962; 1983). the Generalized Bmun Rule (Dean ED-, T.. 1970. erosion during storm conditions: Proceedings. 12th International Conference on : American So- and Maurneyer, 1983), the Edelman method (Edelman, ciety of Civil Engineers. New York. p. 1305-1307. 1970). and various others (e.g., Everts, 1987). The basis EMERY.K. 0.. AND KUHN,G. G.. 1982. Sea cliffs: their processes. pro- for these models is the assumption that the shoreface will files. and classification: Geological Society of America Bulletin. v. 93. retreat landward and upward due to a rise in sea level. Most p. 644-654. of the models assume: (1) a constant wave climate, (2) a EVANS.M. W., HINE.A. C., BELK~NAP,D. F., AND DAVIS.R. A., 1985. Bedrock controls on development: west-central Florida purely sandy shoreface, (3) no longshore loss of sand, (4) coast. in Oertel, G. F.. and Leathennan, S. P.. cds.. Barrier Islands: sand transport by incident waves only, and (5) no sand Marine Geology. v. 63. p. 263-283. transport seaward beyond the shoreface by storms. These EVERTS,C. H.. 1987. evolution in response to a rise in assumptions are never completely valid. Clearly, the slope sea level, in Nummedal. D.. Pilkey. 0. H., and Howard. J. D.. eds.. Sea-Level Fluctuation and Coastal Evolution: Society of Economic Pa- of the surface over which a shoreline will retreat must also leontologists and Mineralogists Special Publication 41. p. 49-57. be a factor; this consideration, however, is not a part of Fox. W. T.. 1985. Modeling coastal environments. in Davis. R. A.. ed.. any current models. Coastal Sedimentary Environments: Springer-Verlag, New York. p. 665- Pilkey and Davis (1987) applied several of these models 705.. --. to the barrier-island coast of North Carolina. South of GRIGGS.G. B., 1987a. California's retreating shoreline: the state of the problem. in Magoon. 0. T.. ed.. Coastal Zone '87. American Society Lookout, the Bmun Rule, Generalized Bruun Rule and a of Civil Engineers. New York. p. 1370-1383. model incorporating the slope of the migration surface all GRIGGS.G. B.. 1987b. The production, transport and delivery of coarse- predicted similar recession for a given rise in sea level. North grained sediment by California's coastal . in Kraus, N. C.. ed.. of Cape Lookout, consideration of the slope of the migra- Coastal '87. American Society of Civil Enginem. New York. tion surface predicted a much greater recession than did p. 1825-1838. GRICl(jS. G. B.. AND SAVOY.L. E.. 1985, Living With the California Coast: Bruun-related models, perhaps because the islands are in Duke University Press, Durham, North Carolina. 393 p. an "out-ofsquilibrium position" with respect to present sea KRAFT.J. C.. CHRUSTOWSKI,M. J., BUKNAP,D. F.. TOSCANO.M- A., level. That is, the recent (since about 4.5 ka) near-stillstand AND FLETCHER,C. H., 1987. The transgressive barrier-lagoon coast of of relative sea level combined with a large sediment supply Delaware: morphostratigraphy, sedimentary sequences and responses has permitted these islands to remain in place, while the to relative rise in sea level. in Nummedal. D.. Pilkey. 0. H.. and back-barrier (Albemarle and Pamlico Sounds) have Howard. J. D.. eds., Sea-Level Fluctuation and Coastal Evolution: So- ciety of Economic Paleontologists and Mineralogists Special Publica- widened. tion 41. p. 127-143. Numerical models of the coastal environment have be- hKHW V. C.. AND m,A. S.. eds.. 1989, Applications in Coastal come important tools for studying shoreline changes. Sev- Modeling: Elsevier Science Publishers. Amsterdam. 387 p. eral approaches to modeling the coastal erivironment are ~THUL!.S. P., 1984. Shoreline evolution of Assateague Island: Shore given by Fox (1985) and Lakhan and Trenhaile (1989). It and Beach. v. 52, p. 3-10. MAY. S. K.. DOLAN.R., AND HAYDEN.B. P.. 1983. Erosion of U.S. is important to note, however, that the coastal system con- shorelines: MS. Transactions. American Geophysical Union v. 64, p. sists of many poorly understood components (e.g., shore- 521-523. face, tidal-inlet, and so forth) that cannot yet be modeled MAY.S. K.. KIMBALL.W. H.. GRADY.N.. AND DOLAN.R., 1982, CEIS: realistically. Wright and others (199 1). for example, dem- The Coastal Erosion Information System: Shore and Beach. v. 50. p. onstrate that much more is involved in shoreface sediment 19-26. MORTON.R. A., 1979, Temporal and spatial variations in shoreline changes transport than just the incident wave field considered in most and their implications, examples from the Texas Gulf Coast: Journal models. It is doubtful that any existing models can predict of Sedimentary Petrology, v. 49. p. 1 101-1 11 1. shoreline-erosion rates with an accuracy useful to coastal NATIONALRESEARCH COUNCIL. 1990, Managing Coastal Erosion: National communities. Academy Rcss. Washington, D.C.. 182 p. NORRIS,R. M.. 1990. Sea cliff erosion: A major dilemma: California REFER WCES Geology. August. p. 171-177. BRUUN.P.. 1962. Sea-level rise as a cause of shore erosion: Proceedings, NUMMEDAL,D., 1983, Barrier islands, in Komar, P. D., ed., Handbook American Society of Civil Engineers, Journal of the Waterways and of Coastal Processes and Erosion: CRC Press, Boca Raton, Florida, p. Harbors Division, v. 88, p. 1 17- 130. 77-121. BRUUN,P.. 1983. Review of conditions for uses of the Bmun Rule of OERTEL.G. F.. LUDWICK.J. C.. AND OERTU. D. L. S., 1989. Sand ac- erosion: Coastal Engineering. v. 7, p. 77-89. counting methodology for barrier island sediment budget analysis. in DEAN.R. G., AND MAURMEYER.E. M., 1983. Models for beach profile Stauble, D. K.. ed.. Barrier Islands: Process and Management: Amer- responses. in Kornar, P. D.. ed.. Handbook of Coastal Processes and ican Society of Civil Engineers. New York. p. 43-61. Erosion: Boca Raton. Florida. CRC Press. p. 151-166. ORFORD,J., 1987. Coastal processes: the coastal response to sea-level DEUAWT. J. M., AND ~MHERMAN.S. P.. 1985. Mainland influence on variation, in Dcvoy. R. J. N.. ed., Sea Surface S~dies:A Global View: coastal transgression: Delmarva Peninsula. in Oertel, G. F.. and Leath- Croom Helm. London, p. 415463. man, S. P., eds.. Banier Islands: Marine Geology, v. 63. p. 19-33. PENLAND.S.. SUER. J. R.. AND BOYD.R.. 1985. Barrier island arcs along DOLAN. R.. ANDERS.F.. AND KIMBALL,S.. 1985. Coastal Erosion and abandoned Mississippi River deltas. in Oenel. G. F.. and Leatherman. Accretion: National Atlas of the United States of America: Reston. S. P.. eds.. Barrier Islands: Marine Geology, v. 63,p. 197-233. Virginia. Department of the Interior, U.S. Geological Survey. 1 sheet. PILKEY.0. H.. AND DAVIS.T. W.. 1987. An analysis of coastal recession DOLAN. R., FENSTER,M.. AND HOLME.S.. 1990a. Erosion of U.S. S~OE- models: Nonh Carolina coast, in Nummedal. D., Pilkey, 0. H., and lines: Gcotimes. v. 35. p. 22-24. Howard, J. D.. eds., Sea-Level Fluctuation and Coastal Evolution: SO- U. S. SHORELINE EROSION

ciety of Economic Paleontologists and Mineralogists Special Publica- WOOD. W. L.. HOOVER.J. A., STOCKBERGER.M. T.. AND ZHANG.Y., tion 41. p. 59-68. 1988. Coastal Situation Report for the State of Indiana: Gnat Lakes SCOR WORKINGGROUP 89, 1991, The response of beaches to sea-level Coastal Research Laboratory. Purdue University. West Lafayette. In- changes: A review of predictive models: Journal of Coastal Research, diana. 190 p. V. 7, p. 895-921. WRIGHT.L. D., BOON,J. D., KIM.S. C.. AND LIST, J. H., 1991. Modes TRENHAILE.A. S.. 1987. The Gtomorphology of Rock Coasts: Oxford of cross-shore sediment transport on the shoreface of the Middle At- University Ress. New Yo*. 384 p. lantic : Marine Geology. v. 96, p. 19-51.